作者: Fei-Fei Li , P. Perona
DOI: 10.1109/CVPR.2005.16
关键词: Caltech 101 、 Dynamic topic model 、 Unsupervised learning 、 Training set 、 Theme (narrative) 、 LabelMe 、 Bag-of-words model in computer vision 、 Machine learning 、 Computer science 、 Visual dictionary 、 Categorization 、 Natural language processing 、 Artificial intelligence 、 Contextual image classification
摘要: We propose a novel approach to learn and recognize natural scene categories. Unlike previous work, it does not require experts annotate the training set. represent image of by collection local regions, denoted as codewords obtained unsupervised learning. Each region is represented part "theme". In such themes were learnt from hand-annotations experts, while our method learns theme distributions well distribution over without supervision. report satisfactory categorization performances on large set 13 categories complex scenes.